diy solar

diy solar

Adding storage to my Enphase system

Despite the hurdles and learning curve of Schneider, this looks great! Good sticking it. Congrats.
You are giving me hope that i will eventually work as I desire.
Buying a second inverter to act as the charger seems like a failure on Schneider's end when the hardware of the XW can do what you want.

Once you're happy with the operation would you share your program? I'd have a much better footing to get it working as I desire if I was starting with something working and if you were willing to help...
 
If I make this work as planned, I would certainly share my code here. For a small fee, I could even assemble the box, make it look nice, program it etc. Maybe sell them through the Will or AltE
 
Making a little more progress. I got the Nano-10 PLC to link to the XW. I have not tried sending commands yet, but the inverter replied with a <Connection OK> message. With a little luck, I should be able to read and write data tonight.

We had a bit less sun the last 2 days, so it was pulling the battery down a bit further. I upped the charge current to 34 amps to get it full before the 4PM shut down and it just made it. 98% 356 amp hours available on the BMS. It actually went into Absorb at 57.6 volts for over 20 minutes. Solar power is already falling more than yesterday, so I also dropped the export power down to just 10 amps at 240 volts. It is going to be crazy hot, I want a little more reserve in the batteries, and I don't want to work them so hard in the heat.
 
I thought I would share this, similar LG cells. But from their resu batt, not from Chevy

I did a discharge from 4.2-3.0 volts per cell.
I don't know if the Chevy Bolt cells would have a similar voltage curve, but I expect they do. The odd angle change at about 3.7 volts, otherwise the voltage is very flat until 3.4 volts.

szpcyx2.jpg
 
Here is my battery charge and discharge curves from 3 days ago. The last 2 days I was messing with settings, this was all on it's own.
XW-batt-9-1.PNG
That is a constant current charge at 30 amps from 8:30 AM, and it hit the Absorb voltage at 3:45 PM. They do have that odd "Knee" with a bit of an "S" curve from 50 volts up to about 52.5 volts. That would be 3.57 to 3.75 volts per cell. In that area, I did see the balancer kick in a few times because the highest cell was 0.008 volts higher than the lowest cell. The Absorb charge only lasted 20 minutes. The batteries only got to rest for 10 minutes

The discharge kicks in at 4 PM. I had it set to output 11.5 amps at 240 volts to the panel, but it only adds to the solar to make the total 12 amps out. At the start, the solar was still putting out most of the power. It was only pulling 3.8 amps from the batteries. That is just 217 watts. As the sun goes down, it ramps up and draws more current to keep 11.5 amps flowing to my main panel. The current trace shows the wiggles as the XW is constantly adjusting the current to control the grid output based on how much power is coming in from the solar. By 5:30 PM The XW was pulling 1300 watts from the battery at 55.8 volts and 23.7 amps. As more of the power is from battery than solar, the current trace starts getting smoother. The XW is now more in control. The solar output quit completely at 6:53 PM. The XW was then pulling 53.1 amps at 52.8 volts, just over 2,800 watts. From 7 PM to 9 PM the current just slowly increased as the battery voltage was dropping to hold that 2,800 watts. Just before 9 PM I see 49.33 volts and 57.8 amps for 2850 watts. The battery voltage bounces back up a bit and settles to 49.9 volts after an hour.

With my current settings, I don't plan to pull it quite that far down, I want to stay above 50% charge, this was more like 38-40% charge with each cell at 3.52 volts. That s not a bad level to let them rest the night.
 
Nice
Interesting that both show the knee, but I suppose that does make sense.

Funny, you charged 6p at 30 amps for most of the day. My little test pack is 1p and took 30 amps about 2 hours, but that was through significantly more of the voltage range.

I'm hoping I can figure out the inverter stuff so I can order a bolt pack. Your set up is making me jealous.
 
My 6P is rated for 360 AH total, so the 30 amp charge rate is just C/12. If the pack was completely dead and charged to full, it would take over 12 hours at that rate. This charged for 7 hours. 7 x 30 = 210 AH. The BMS calculated it was at about 40%, so had to put 60% back in. 210/0.60 = 350 so I think it is pretty close. I am still stopping short of completely full by about 5-10%. Then numbers seem to work, and it looks like a full round trip is only wasting 5-7% of the energy. That is more efficient than I was expecting.
 
I have been at my PC working on coding the PLC, and it is making progress. I now having it using the real time clock to update things, so I can certainly make it do functions based on time of day. Now I need to try and read registers in the XW-Pro. Read command should be safe and not mess anything up. I'll try watching battery current and see how well it matches the Gateway Web UI.

Last night I had lowered my export power a little so it left the battery at 50.5 volts. And I had also raised the charge current a bit, so it was running at almost 35 amps this morning. The XW-Pro went into Absorb charge at 57.6 volts. I like the custom setting better, it actually matches the voltage I had set for "Absorption Voltage Set Point". It went into constant voltage and the current slowly ramped down from 30 amps to under 2 amps over a 90 minute period, and was still putting in about an amp for over another hour. The time settings stopped all charging at 3:55 PM. At that time, the BMS reported the battery at 99% charged with 359.9 AH available, even though it is just at 57.2 volts now in the BMS. The individual cell voltages average 4.086 volts, and the lowest is 4.084 volts and the highest is 4.087 volts. The battery temp sensors are showing 34C and 35C. That is 95F which sounds a little high, until you realize the air temp outside of my non cooled garage is 110F in the shade. The area has declared a heat emergency, and we may have rolling blackouts. And they are predicting even hotter tomorrow, with temps up to 119F (48.333C) expected. My A/C has not stopped. And the solar panels are hot and the sky is hazy, so solar production is a bit down also.
 
This heat wave SUCKS!
It topped 113F today, and it will be hotter tomorrow. I just got my SCE usage for yesterday. OH CRAP!
I still had to buy over 45 KWH's to keep the A/C cranking all day. Topped out pulling over 3.5 to 4 KW for 3 full hours. When the battery started pushing, it did hold it under 2.5 KW for the 5 hour long peak rate window though.
 
We have topped 108F before noon today, on to 117 predicted now. And also predicting 114 again for Monday. The panels are so hot they are down about 10% on power from yesterday, which was also hot. The only saving grace is the low humidity. In my garage right now, the air temp is at 95F or 35C which is not too bad for the battery bank and the inverter, but it is climbing. Charging at 34 amps, the battery bank is still cooler than the air in the garage at 34C (93F).

I just may be testing the backup power today. We have had several warnings of possible rolling blackouts due to the heat. I don't have my load circuits moved over yet, but if the power goes out, it will be easy to measure which circuits need to go, and the wires are right there waiting.
 
No power failure yet, but as the sun is falling, it is still 105 out and everyone is running the A/C and all the solar panels are starting to shut down. Now comes the real grid test.

I was busy around the house and zoned out, so I didn't change the XW settings until almost 5 PM. The solar was still doing ok, so when I did get it online, it was only pushing 1,000 watts. It's now almost 6:30, the solar production is down to just 200 watts, and the battery is pushing 2,300 watts. With the A/C running, I am probably pulling about 3,000 watts from the grid.

I have run into a bit of a stumbling block. I have an e-mail into Triangle Research. The PLC is connecting to the XW ans it is seeing the connection, but when I ask for data, I am not getting a reply. No error, nothing. I think it is only sending a 16 bit word request, and the XW response is 32 bit. So the XW is just ignoring it. The command must be for something else. I have a few other things to try, but I was hoping to hear back from Triangle to see if they have an example code for 32 bit read and write to Modbus TCP. There are a few 16 bit values in the XW, I am ging to try and read those tonight and see if I can make some more progress.
 
Is what you are doing with the gateway and programming so you can communicate with the BMS to the charge controller?
You have so many things going on and much of it I don't understand so it is hard for me to be sure which piece of the puzzle you are working on.
 
...topped 108F before noon today, on to 117 predicted now...
That's some pretty serious temperatures... what's up with that? Hate to think what would happen if the grid did fail there.
Stay safe!
 
He lives near the lake of fire. :devilish:
Solar salespeople and solar people in general often say it's not worth the extra cost to have power when the grid goes down.
There is no way in hell I will set up a solar system for my home that won't operate when the grid is down and the sun is shining.
Hope your system is keeping you safe. Seems like you should get a break soon. It's getting late in the year.
 
ArthurEld,

Yes, I have a lot going on. The main reason I am working on having a PLC (programmable logic controller) talk to my Schneider XW-Pro inverter is so that I can have it properly change modes between charging when the solar is making extra power, and invert from batteries to help reduce how much power I use from the grid during the evening peak rate time of use which is from 4 to 9 pm here with SCE. The real problem (in my opinion) is a programming error by Schneider. With the settings I am using now, it all works, except I need to change two values to force it to either charge or discharge based on the battery voltage. If the voltage is above the two numbers, it will only discharge/invert, and if the battery voltage is above the two numbers, it will only charge. These numbers should be time dependent when it is blocking charging or blocking grid power output, it should ignore that value, but it does not. If the battery is below the recharge volts, even if charging is blocked, it can't go into invert, and if the battery voltage is above the grid support voltage, it can't go into charge, even if grid support is blocked by time. So my step one is just having the PLC change those values based on time of day to make it charge when my solar is producing, and invert during peak time.

Once I get that part working, I can add more functions. I can actually read how much power is being produced and used and adjust the charge rate and grid sell power numbers to "Flatten the curve" to use as little grid power as possible while not exporting power to the grid to keep it in the battery. If I ever do find the details of the JK BMS RS-485 data, I could also add that and have the PLC report the state of charge and such, but that would be much further down the road.

svetz, Ampster,

Yes, this heat wave is crazy. The good thing is that the weather has been predicted a week in advance, so people have been prepared for it. The "Lake Fire" is "under control", Castaic Lake even opened the beaches. We have a few other small firs in a 30 mile radius, but nothing too serious right now. There is a bit of smoke and haze and air quality is rated as poor/unhealthy. And I have more soot and ash on my solar panels again.

They did manage to keep the grid power up last night. I am sure they had to crank up some "peaker plants" and maybe dump some resevoir water down through turbines to make it up, but the grid held on to run all the A/C running in the valley as the thousands of solar grid tie systems went offline about 7 pm last night with the outside temp still over 105F. I did what I could, I pushed over 8 KWH's out of my battery as the sun dipped and kept my draw from the grid down to under 3 KW's all night. Without the battery, it would have been pulling over 5 KW's to keep the A/C blowing cold.

To zero out my grid with the A/C running, I would need about double my battery capacity. The Schneider inverter will take it. But to move that much power safely, I would certainly have to have a WattNode reading my grid current and carefully adjusting the export current as the A/C compressor cycles. The 3,600 watt swing is more than I should be exporting, so when it stops, I need to back off the inverter very quickly.

Obviously my current 16 solar panels can't make enough power on their own to make that back up, but I could have the batteries charging from grid at a slow rate all night long and not increase my purchased power. But by the letter of the NEM 2.0 agreement, grid power can't be used to charge and sell back to the grid. BUT... If I don't export any power and it is all used to run my home, I am not violating the rules. And in many ways, I am helping the utility by using extra power when demand is low and not taking power when the demand is high. They certainly should not complain, and hopefully in the future, we will get credit for doing our part to help the grid. I have room to add about 6 more solar panels fairly easily. But even that can't keep my whole home A/C alive in these crazy heat conditions. 2 days ago, the latest I have from SCE, it did top 100, but not like yesterday, and even with my solar making 24 KWH's, I still had to buy 50 KWH's from SCE. That would take a serious 12.5 KW solar array to make 74 KWH's of power in 6 sun hours. I just need 200% more panels. I could cut my consumption a little if I had to, but not that much. And before anyone starts whining, my thermostat was up to 80. With it being over 110 outside, the indoor house temp went up 2 degrees with the A/C running flat out. I have excellent thermal insulating windows and doors, and the attic and walls are well insulated too. This kind of heat is no joke. Even for being halfway between Hollywood and the Mojave Dessert, this is not normal. We only get conditions like this a few days a year. But when they do come, you have to be ready for it. It is a "Dry Heat" so it may not feel quite as bad as 100 back north of Chicago, but it still feels wrong. The inside of my garage did get to 100, but I had to shut off my vent fan, as that made it worse fast as the air exhausted was being replaced by the even hotter air outside. Same with the attic fans. Mine normally turn on at 90 degrees, but it makes them pull in 115 air! So another future project will be adding smart control to only run the attic and garage fans when they are actually hotter than the outside air. Not just on inside heat alone.
 
Well this is not good. I look outside and the sky is getting a bit grey. I open the door and I can taste the ash in the air. We have a new fire burning up wind from us. From what I can find on the news, a car coming down the Interstate 5 from the Tejon pass caught on fire. They pulled to the side of the road and started a brush fire. The fire is bad enough, they are doing helicopter water drops and they had to close the southbound I5 coming down from the pass. The northbound lanes are almost a half mile away, so they are still open. Hopefully they get this fire out quick as it is in a bad place, and the smoke is covering the whole valley now. The ash in the air has my solar output cut by 30% from yesterday. No clouds in the sky, just smoke and ash.
 
I have several big paper bags set aside just for tossing stuff into - and my insurance folders and family photos in a box. I've kept them all handy since two fires broke out less than a 1/4 mile from me a couple of weeks ago. Lightning and carelessness. Prepare for the worst - and hope for the best. Times like these I'm grateful to be totally off-grid... PG&E has Grass Valley on the cutoff list for tonight due to high winds and a red flag warning. Everybody stay safe out there. Best.
 
They got the i5 brush fire out, only burned 4 acres, but it was some of the worst smoke we have had here. Wish you all well too. There are some tough times. The heat and dry weather makes all the brush ready to go in a big hurry.

The big 7,000+ acre fire near Yucaipa, CA was started by a "Gender Reveal Party". They had a smoke bomb with colored smoke to show if it was a boy or girl, well, it lit the hillside on fire. Stupid might not be illegal, but COME ON!!
 
ArthurEld,

Yes, I have a lot going on. The main reason I am working on having a PLC (programmable logic controller) talk to my Schneider XW-Pro inverter is so that I can have it properly change modes between charging when the solar is making extra power, and invert from batteries to help reduce how much power I use from the grid during the evening peak rate time of use which is from 4 to 9 pm here with SCE. The real problem (in my opinion) is a programming error by Schneider. With the settings I am using now, it all works, except I need to change two values to force it to either charge or discharge based on the battery voltage. If the voltage is above the two numbers, it will only discharge/invert, and if the battery voltage is above the two numbers, it will only charge. These numbers should be time dependent when it is blocking charging or blocking grid power output, it should ignore that value, but it does not. If the battery is below the recharge volts, even if charging is blocked, it can't go into invert, and if the battery voltage is above the grid support voltage, it can't go into charge, even if grid support is blocked by time. So my step one is just having the PLC change those values based on time of day to make it charge when my solar is producing, and invert during peak time.

Once I get that part working, I can add more functions. I can actually read how much power is being produced and used and adjust the charge rate and grid sell power numbers to "Flatten the curve" to use as little grid power as possible while not exporting power to the grid to keep it in the battery. If I ever do find the details of the JK BMS RS-485 data, I could also add that and have the PLC report the state of charge and such, but that would be much further down the road.

svetz, Ampster,

Yes, this heat wave is crazy. The good thing is that the weather has been predicted a week in advance, so people have been prepared for it. The "Lake Fire" is "under control", Castaic Lake even opened the beaches. We have a few other small firs in a 30 mile radius, but nothing too serious right now. There is a bit of smoke and haze and air quality is rated as poor/unhealthy. And I have more soot and ash on my solar panels again.

They did manage to keep the grid power up last night. I am sure they had to crank up some "peaker plants" and maybe dump some resevoir water down through turbines to make it up, but the grid held on to run all the A/C running in the valley as the thousands of solar grid tie systems went offline about 7 pm last night with the outside temp still over 105F. I did what I could, I pushed over 8 KWH's out of my battery as the sun dipped and kept my draw from the grid down to under 3 KW's all night. Without the battery, it would have been pulling over 5 KW's to keep the A/C blowing cold.

To zero out my grid with the A/C running, I would need about double my battery capacity. The Schneider inverter will take it. But to move that much power safely, I would certainly have to have a WattNode reading my grid current and carefully adjusting the export current as the A/C compressor cycles. The 3,600 watt swing is more than I should be exporting, so when it stops, I need to back off the inverter very quickly.

Obviously my current 16 solar panels can't make enough power on their own to make that back up, but I could have the batteries charging from grid at a slow rate all night long and not increase my purchased power. But by the letter of the NEM 2.0 agreement, grid power can't be used to charge and sell back to the grid. BUT... If I don't export any power and it is all used to run my home, I am not violating the rules. And in many ways, I am helping the utility by using extra power when demand is low and not taking power when the demand is high. They certainly should not complain, and hopefully in the future, we will get credit for doing our part to help the grid. I have room to add about 6 more solar panels fairly easily. But even that can't keep my whole home A/C alive in these crazy heat conditions. 2 days ago, the latest I have from SCE, it did top 100, but not like yesterday, and even with my solar making 24 KWH's, I still had to buy 50 KWH's from SCE. That would take a serious 12.5 KW solar array to make 74 KWH's of power in 6 sun hours. I just need 200% more panels. I could cut my consumption a little if I had to, but not that much. And before anyone starts whining, my thermostat was up to 80. With it being over 110 outside, the indoor house temp went up 2 degrees with the A/C running flat out. I have excellent thermal insulating windows and doors, and the attic and walls are well insulated too. This kind of heat is no joke. Even for being halfway between Hollywood and the Mojave Dessert, this is not normal. We only get conditions like this a few days a year. But when they do come, you have to be ready for it. It is a "Dry Heat" so it may not feel quite as bad as 100 back north of Chicago, but it still feels wrong. The inside of my garage did get to 100, but I had to shut off my vent fan, as that made it worse fast as the air exhausted was being replaced by the even hotter air outside. Same with the attic fans. Mine normally turn on at 90 degrees, but it makes them pull in 115 air! So another future project will be adding smart control to only run the attic and garage fans when they are actually hotter than the outside air. Not just on inside heat alone.
I have been following your progress but a lot of it is over my head. And the time of use thing doesn't apply to me at this time since my power company is net zero. I noticed you commenting in some of the BMS threads and I figured connecting to the controller would be something you are interested in. It seems so complicated. I guess letting the BMS cut the power isn't so bad.
I just realized you are in California. I lived in Los Angeles beach area for 20 years. I miss it a lot.
 
Back
Top